In the context of business, interactions between individuals generate social systems that emerge anywhere within a corporation or in its relations with external agents. These systems influence the behaviors of individuals and, as a result, the collective actions we usually attribute to corporations. Social systems thus make a difference in processes of action that are often morally evaluated by internal and external agents to the firm. Despite this relevance, social systems have not yet been the object of specific attention in the literature on moral agency in business. To fill this gap, I construct a theoretical framework based on Luhmann’s ideas on social systems and morality. In particular, I argue that morality is a phenomenon that occurs in communication, and an agent can engage in morality in three different ways (as a moral factor, an expressive moral agent, or a reflective moral agent) depending on the functions it plays in communicative interactions. Based on this framework, I argue that social systems in business can be considered moral agents of those types.
Notes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Introduction
In the literature on moral agency in business, it is frequent to identify two groups of authors, depending on the subject to which they attribute moral agency (e.g., Mulgan, 2019; Rönnegard & Velasquez, 2017). One group of authors—usually called individualists—holds that only individuals can be the subjects of moral expectations and responsibilities (Mansell, 2008; Velasquez, 1983). Another group of authors—usually called collectivists—argues that firms can also act with moral agency (Hess, 2014; List & Pettit, 2011). Each position has its merits. Individualist authors stress that only individuals can make decisions in an embodied and rational way. Collectivist authors stress that firms apply formal structures to make decisions that cannot be assigned to individuals. However, the possibility of attributing moral agency to individuals or firms leaves a relevant gap: in many situations, the decisions and behaviors of individuals are strongly conditioned by social dynamics that are local and stable. Sometimes, these dynamics occur within the perimeter of the firm. It would be the case of an informal group of engineers sharing work routines within a department. Other times, these dynamics occur in interactions beyond that perimeter. It would be the case of a team of consultants collaborating with employees of a client company. In any of these cases, social dynamics play a role that cannot be passed to individuals or corporations.
In this essay, I attempt to fill that gap by answering two questions: what conditions should an entity of any nature meet to be recognized as a moral agent? And do the social systems that emerge in business meet those conditions? To address these questions, I assume a systems approach. Following Niklas Luhmann’s ideas on social systems (1995a, 2006) and morality (1991, 1992a, 1996), I construct a theoretical framework for understanding social systems and their connection to morality. Specifically, I define a social system as an entity that emerges from the communicative interactions between various agents linked in a relationship of mutual contingency. I also interpret morality as a dimension of communication that enables the agents involved in a social system to manage their interdependencies through the index of esteem. Following Luhmann (1995a), I interpret esteem as “a generalized recognition and evaluation which honors the fact that others accord with the expectations one believes must be assumed for social relations to continue” (p. 235). Esteem thus expresses—as a symbol—the degree of alignment that any agent appreciates between its expectations and those of another agent with whom it is involved in a social interaction. From this framework, I answer the first question by distinguishing three types of moral agency (moral factor, moral expressive agency, and moral reflective agency) according to the abilities of the entity to express and receive esteem in its relationships with others, and to communicate about the rules and judgments connected to esteem. I address the second question by applying these categories to the social systems that emerge in business. As a result, I will show that those systems should be considered moral agents when they fulfill the appropriate functions in interactions.
Advertisement
My proposal is located at the convergence of two lines of research: the philosophical inquiry on moral agency in business and the literature on the communicative constitution of organizations (CCO), which includes Luhmann’s theory (Cooren & Seidl, 2022). The former has directly addressed the question of moral agency in business, but has neglected the role of communication. The second has not yet assumed moral agency as a relevant issue. This location allows me to enrich both lines of research. In relation to the study of moral agency in business, I emphasize the importance of communication for morality and present social systems as adequate moral agents. In relation to the CCO approach, I develop a communicative perspective to explain moral agency in business.
This essay is structured as follows. First, I review the literature addressing the issue of moral agency in business. To do so, I identify the main approaches according to their understanding of communication. Second, I present my framework to understand social systems. Specifically, I explain how these systems arise, what agents may contribute to their existence, and how we can identify them in business. To build this framework, I take as a basis Luhmann’s ideas on social systems and morality, which I correct and complete in some relevant points. Third, I state the conditions for moral agency. From these conditions, I distinguish three types of moral agency. Fourth, I assess whether the social systems that emerge in business meet the conditions for each type of moral agency. Finally, I point out my main contributions and several lines for future research.
Moral Agency and Communication: Theoretical Approaches
Communication as a Peripheral Phenomenon to Moral Agency
In the early days of philosophical reflection on moral agency in business, authors focused on the conditions a firm should meet to be considered a moral agent (Donaldson, 1982; French, 1979). In a definition generally accepted by collectivists, Pettit (2007, p. 175) claimed that an entity of any kind (a human being, an animal, a robot…) would be a moral agent if it acted with autonomy, normative judgment, and self-control. Based on this image, collectivists have argued that firms have internal structures that enable them to meet those conditions, so they should be considered moral agents (Björnsson & Hess, 2017; Bratman, 2022; List & Pettit, 2011). By handling arguments in this way, collectivist authors assume a functional approach: moral agency depends on an entity fulfilling certain conditions that become noticeable when acting, whatever internal attribute that entity employs to achieve it.
The functionalist approach has been dominant on the collective side. However, other collectivist authors have applied a pragmatic approach to recognize corporations as moral agents (Hsieh, 2017; Hussain & Sandberg, 2017). Manning (1984) made this turn explicit by stating that “corporations are not persons in the same way that you and I are […]. We want to modify their behavior if we think it inappropriate. In deciding how to do this, the considerations are utilitarian” (p. 83). According to this logic, by recognizing firms as moral agents, we make them the object of our expectations and judgments, we grant or withdraw our material and symbolic support and, as a result, we offer their managers an incentive to modify their behavior. As a recent example, Mulgan (2019) explores the moral agency of corporations in a plausible future aggravated by climate change and material scarcity, in which digital beings coexist with human individuals. Given such a potential scenario, Mulgan underlines the utility of attributing moral agency to corporations, as these collectives have a great capacity to shape the future. Pragmatic collectivists reinterpret the nature of moral agency. Instead of seeing it as a functional attribute of corporations, they see it as a device that allows society to control their behaviors. Despite their differences in arguments, functional and pragmatist collectivists agree on a relevant idea: corporations are moral agents analogous to individuals.
Advertisement
As a reaction to the collectivists’ arguments, some authors have defended that only individuals can be considered moral agents in business (Rönnegard & Velásquez, 2017; Velásquez, 1983, 2003). These authors have applied a common strategy: they highlight some conditions that firms do not meet but individuals do. From this point, they claim that firms are mere abstractions and only individuals should be considered moral agents. Similar to functional collectivists, these authors take the prototypical human being as a reference, but assume a phenomenological approach to underline some attributes that only embodied beings, endowed with consciousness and moved by emotions, can experience (Mansell, 2008; Sepinwall, 2017).
The distance between individualists and collectivists has led many to identify two sides in the debate over moral agency in business (e.g., Mulgan, 2019). This opposition, however, deserves comment. Despite their differences, authors on both sides share two crucial ideas: individuals and corporations are the only plausible agents for morality in business, and communication is interpreted as a process to convey messages without interfering in moral dynamics. Regarding the first point, research on moral agency in business is often framed from the question: are corporations moral agents analogous to individuals? (e.g., Pettit, 2007 on the collectivist side, and Sepinwall, 2017 on the individualist side). By stating the question this way, authors offer their arguments for answering affirmatively or negatively. The result is a bifurcated field, with opinions supporting one of the sides. Regarding the second point, it is often admitted that agents use communication, but it is not seen as an essential factor for moral agency (see critically, on both sides, Donaldson, 1982 and Velasquez, 2003). In contrast, several authors have recently highlighted the role of communication when it comes to attribute moral agency in business. In the following subsection, I introduce these proposals.
Communication as a Central Phenomenon to Moral Agency
In recent years, several authors have argued that communication is an essential dimension of interactions that must be considered to explain moral agency in business. Prominent examples of this communicative turn are French’s (2017) diachronic responsibility, Pettit’s (2017) conversable firm, and Cooren’s (2020) notion of ventriloquism applied to business ethics. These proposals are still marginal and lack a theoretical thread connecting them. Indeed, this essay has the additional value of locating them together in the field of theoretical research on moral agency in business.
French (2017) maintains his emphasis on the corporate internal decision structure (CIDS), but takes a step forward to propose a narrative approach to moral agency, taking into account the self-reference articulated by firms “in annual reports, in advertising, in legal documents, in internal and external statements of corporate culture, and in policies” (p. 62). In French’s proposal, moral responsibility has two dimensions: a synchronic responsibility, linking the firm’s recent behaviors with its current CIDS, and a diachronic responsibility, which connects the firm’s past behaviors to its current self-narrative (this narrative being an expression of its current CIDS). French suggests that diachronic responsibility for past wrongdoing depends on the degree to which the firm’s current self-narrative remains consistent with the CIDS that caused that behavior.
In contrast to French, Pettit (2017) interprets communication as a dialogical phenomenon, rather than a monological one. In particular, Pettit describes firms as conversational agents that behave and speak through authorized representatives. Pettit maintains his functionalist approach of the past, but now interprets communication as an essential dynamic for exercising agency. Through language, firms apply their capabilities to hold beliefs, express intentions, and make promises. A firm’s functional ability to speak enables it to commit, and other actors can criticize the firm when its behaviors do not conform to the expressions of its beliefs and intentions.
Cooren’s proposal (2020) is far from those of French and Pettit. If the latter are philosophers of moral agency who address the question of communication, Cooren is a communication theorist who addresses the question of moral agency. According to Cooren, moral agency arises in the context of a collective dialog on a topic in which human beings give voice to other human and non-human entities. Moral agency is interpreted as an exercise of ventriloquism (Cooren, 2016), as human participants can act as the medium for the relevant elements of the situation— “facts, principles, future generations, ecosystems…” —to express and guide consensual decisions (Cooren, 2010, p. 176). Moral agency is now interpreted as a distributed phenomenon involving human and non-human agents with the capacity to make a difference in a communicative situation.
These proposals highlight something that went unnoticed in previous research on moral agency in business: communication is a phenomenon that should be considered when identifying moral agents. Each of these authors emphasizes one type of agents that participate in communication—corporations in the case of French and Pettit, and individuals for Cooren—but they do not recognize communication as having any ontological force to give existence to social systems. In the next section, I attempt to fill this gap. In particular, I construct a theoretical framework that describes social systems as entities that emerge in communication.
Social Systems and Social Agency
Social interactions are often described as a positive bond shared by an extended group of human beings (Schiefer & Van der Noll, 2017). This idea is challenged—at least partially—by Luhmann, who replaces human beings as the primary element of those interactions with communication. In this section, I construct my image of social interactions. To do so, I take Luhmann’s ideas, which I correct and expand on some relevant points.
Emergence of Social Systems
Luhmann rooted the emergence of social systems in the interactions between individuals. Following his ideas on this point (Luhmann, 1992b, 1995a Chap. 3, 2002), when two actors (A and B) begin to interact, a process of communication starts. In this process, both actors are opaque to each other, as neither can directly observe the internal states of the other. This opacity burdens the relationship with an uncertainty that the actors can only manage through the mechanism of mutual expectations. In order to communicate with each other, A needs to form expectations about the behavior of B, and B needs to form expectations about the behavior of A. For example, the teacher expects the student to listen to his lesson, and the student expects the teacher to explain things he did not know. In addition, each actor needs to infer the expectations that the other has formed about their behavior. The teacher expects the student to expect him to explain his ideas clearly, and the student expects the teacher to expect from him an attitude of attention and respect. By linking this way, both actors transform unmanageable uncertainty into a situation of double contingency (Luhmann, 1990, p. 45; 1995a, p. 108). In this situation, neither actor knows in advance how the other will behave, but their expectations narrow the range of the other’s behaviors that they take as the basis for their own decisions.
If both actors continue to interact, a social system emerges, formed by the communications between those actors (Borch, 2011, Chap. 2). If we look inside the system, we identify a series of recursively linked communications. If we look outside, we realize that a boundary has emerged, separating the communicative dynamics between the actors and their environment (Hernes & Bakken, 2003). As the teacher and the student are linked in a system, their mutual interdependence increases, and that interdependence is reinforced by their communications. The student asks the teacher for homework to do at home. The next day, the teacher asks the student for the homework, checks it, and congratulates the student. This boundary is an essential element of the new social system (teacher-student) and is confirmed in every communication between the actors (Luhmann, 2006). Implicitly but effectively, the reference to the system as an entity makes the actors interpret everything else as external and alien to the system (Seidl, 2005; Weick, 1977). When the student’s parents criticize the school’s teachers for their lack of commitment, the student argues that this opinion is not, from his point of view, applicable to the teacher with whom he participates in a differentiated social system (Fig. 1).
In his extensive work, Luhmann linked social systems to the communicative processes that occur face-to-face between individuals (e.g., 1990, p. 88; 1995a, p. 193). However, the rise of digital communications has proven that this image of social systems is incomplete. Through digital media, it is possible to maintain stable interactions that do not require physical presence or synchronicity (Berglez & Hedenmo, 2023). This would be the case, for example, of a group of friends who organize their vacations through a social network. The actors do not need to be physically present in order to communicate. In this sense, my approach assumes Luhmann’s basic premise (social systems always emerge in communication), but frees the notion of social system from the requirement of verbal and direct interactions. In my framework, a social system exists if recursive communications between two or more actors assume and confirm the boundaries of the system, whether or not all the actors are physically present. The key point is that it is possible to identify a process of recursive communications that links a group of actors, regardless of when, how, or from where those actors communicate (Fig. 2).
Interactions between physically present and non-present actors are frequent in business. Imagine, for example, that a warehouse manager informs one of his subordinates that all employees will have to work overtime during the week to handle an important order he has just received from a customer. Within a few minutes, that employee communicates the boss’s instruction to all his co-workers through a social network. From that moment on, messages flow between employees. A communicative process operates whose boundary is that of the group of employees, even if none of them is physically present. All the workers are aware of the system’s perimeter, assuming it and reinforcing it in their communications. The past interactions between them gave rise to a system, and now, stimulated by the boss’s instruction, communications are flowing again, following the patterns already set for interaction.
Social systems are made up of operations—in particular, communicative events—, and these operations are conditioned by structures produced by operations. Operations and structure are mutually connected, but this relationship of reciprocal influence is not a closed circle. Social actors always communicate in a context of uncertainty—or mutual contingency—, and this uncertainty opens communication to new developments. When two actors communicate, neither of them knows exactly what the other intended to express or how the other has interpreted their message. In such a relationship, the structure is made up of the reciprocal expectations that help the actors to link their recursive communications. Each communication is conditioned by the previous structure of expectations, and that structure will evolve as the actors exchange their messages. The workers expect the warehouse manager to give orders, the manager expects the workers to follow his instructions, and the workers begin to communicate about the possibility of disobeying their boss. The structure of mutual expectations guides communications between the actors, and that structure evolves as communications introduce new elements.
At this point, it is useful to clarify the ontological status of social systems. As Luhmann repeatedly emphasized, social systems only exist when they are recognized by an observer (1990, 1992b, 1995a). The social world is extremely complex, as communications are flowing continuously in many directions. Any person talks to many others in the course of a day, and in any densely populated space—such as a hotel or airport—many communications flow at the same time. Such complexity makes it necessary for an observer to draw “the distinction between system and environment” (Luhmann, 1995b, p. 52). Social reality exists, but it is inaccessible directly and can only be known when an observer distinguishes between what is inside and outside a system (Borch, 2011; Seidl, 2004). When making that distinction, the observer may be participating in the recognized system or may not be participating in it. In the first case, that identification may be reintroduced into the system’s communications, reinforcing the self-referentiality of the system. This would be the case of a family whose members exchange some comments while visiting a museum. In the second case, the observer can use this distinction, without contributing to the maintenance of the system. This would happen, for example, when a mother warns her daughter that she is hanging out with people who may be a bad influence. In both cases, it is important to emphasize that social systems do not appear as well-formed, autonomous and differentiated entities. A social system only comes into existence when an observer recognizes it and confirms its existence through its communications.
Social Systems and Social Agency
Social systems are formed by recursively linked communicative events. According to Luhmann (1992b, 1995a, 2002), these events are performed by individuals (or ‘psychic systems,’ in his terminology) who attempt to manage the uncertainty that arises in their double-contingency relations. When we look at our watch to know the time, we do not believe we are in a situation of double contingency. We accept the time or reject it, if we believe that the watch has no batteries or is broken. On the other hand, when we go to the doctor, we do not know what he will tell us, but we expect him to give us some relevant information as patients. For Luhmann, social systems are made up of communications and not of individuals (hence his radical posthumanism), but only humans can manage their contingency links through symbolic language (hence his residual humanism). From this point, we can address a relevant question: which actors can carry out the communication events from which social systems emerge? Or, to put it differently, which entities may be considered social agents? My answer to this question will lead me to depart significantly from Luhmann’s ideas, opening communication and social agency to a wide range of non-human entities.
Who or What Can Contribute to the Emergence of Social Systems
According to Luhmann, social systems emerge from communications with which human actors attempt to manage contingency situations. Following Luhmann, I accept that social systems emerge from these situations. In contrast, I do not accept that humans are the only entities with the capacity to communicate and manage contingency in such situations. Specifically, I open the notions of sociality and communication to any kind of entity—human or non-human, known or unknown, existing or yet to be invented—that is capable of linking with others in a relationship of mutual expectations, managing the contingency of that relationship through communication. Like Luhmann, I assert that social systems emerge from communication. Unlike him, I assume a strictly functional approach, not taking the human being as the undisputed point of reference. From this point, I define a social agent as any human or non-human entity that is capable of effectively engaging in recursive communications with other entities (of the same or different nature) to manage their double-contingency interactions. In my view, social agency designates the capacity of an entity to contribute to the communications that form a social system, this capacity always being appreciated and attributed by an observer. In recent decades, for example, a large group of researchers have recognized the capacity of certain species of animals (de Waal, 2009; Whiten, 2021) and even plants (Baluska, 2006) to communicate recursively and form complex social systems.
This functional image of social agency implies two main differences from Luhmann’s. First, it dissociates communication from humans. In my view, communication consists of any exchange that allows agents to coordinate their expectations, regardless of the nature of the agents and the code they use to express and interpret messages. It is well established, for example, that many bird species coordinate their interactions during flight using their body positions (Bajec & Heppner, 2009). Second, this approach frees us from making conjectures about what happens inside agents. Notions such as rationality, self-consciousness, or intentionality are not relevant to appreciate sociality. What really counts is how an agent behaves, and not what are the internal mechanisms or processes that lead that agent to act in that way. In summary, I assert that the question about who or what is a social agent cannot be answered a priori and for any ontological category of agents. This question must be addressed for each specific situation, taking into account the functions performed by the entities involved.
Within my general framework, in this essay I focus on the social systems that arise from human communications in business. In particular, I will explore social systems that emerge directly from communications attributed to humans or from communications attributed to certain social systems that emerge in business (informal groups, alliances, corporations, etc.). However, it is important to reiterate that social systems always arise from communications between human or non-human entities that may functionally engage in double-contingency interactions. In some cases, these interactions take place between individuals. In other cases, these interactions involve entities other than individuals. In this line of thought, it is well known that many species of animals (birds, insects, mammals…) and plants maintain strong communicative links between members of a group (Baluska, 2006; Whiten, 2021). Some software programs have also recently begun to engage in recursive communications (Jackson & Williams, 2021). As an illustrative example, in the 2016 Go tournament between human world champion Lee Sedol and AlphaGo (an AI agent developed by DeepMind), each player communicated their intentions and expectations with their actions on the board (Esposito, 2017). During the games, recursive communications between the human and the AI agent formed a social system. In my framework, what counts for social agency is not the nature of an entity, but its functional capacity to engage in recursive communications that generate social systems.
Social Systems as Social Agents
Social systems have no mouth, eyes or ears, but they can interact in a delegated way through other human and non-human agents that act or speak on their behalf. In the field of organizational communication, the authors of the Montreal School have been explaining this phenomenon for several decades (Taylor & Robichaud, 2004; Taylor & Van Every, 1999). In particular, Taylor (2001) explained that organizations act and communicate through human or non-human agents “in imbricated sequences of actings-for,” and organizations thus exist as “an embedded, encapsulated set of agent-instrument linkages relegated to an out-of-consciousness infrastructure” (p. 281). According to Taylor, we attribute an action to an organization, even if that action was not directly performed by that organization. To develop this idea, Cooren and colleagues have convincingly argued that organizations can intervene in social life through an expressive modality they call ventriloquism (see Section “Communication as a Central Phenomenon to Moral Agency”). This notion helps us understand that social systems can make themselves present and interact in contingency relations through texts and human beings speaking on their behalf (Cooren, 2010; Cooren & Sandler, 2014).
The idea of mediated communication can be applied to formal organizations, but also to any other type of social system. A football player may speak to the media on behalf of his team, and a student can address the teacher to express the opinion of her classmates. Social systems are communicatively closed systems—as they are formed by self-referential communications—, but they are also open systems, since they can process information about their environment and interact with agents that are part of that environment (Borch, 2011). We intuitively accept this capacity when we attribute collaborative or rivalry relationships to governments, international institutions, firms, activist collectives or informal groups, among others. We may say, for example, that a company’s top management team has a smooth relationship with the workers’ unions, that a corporation collaborates with its suppliers to develop some joint innovations, or that a commercial department’s performance has been penalized by the conflicts among its informal groups. In all these cases, we assume that social systems of different levels behave as social agents. In all these cases, we understand that social systems interact in situations of double contingency in which a new, larger social system emerges.
In the organizational literature, several lines of research have partially supported the capacity of social systems to act as social agents. In particular, the literature on alliances has extensively highlighted how firms participate in stable relationships with other agents. Corporations often engage in alliances—developing supra-organizational social systems—to implement responsible practices (Lin, 2012), execute operationally complex projects (Lavikka et al., 2015), or innovate (Bustinza et al., 2019). In another line of research, some authors have developed the notion of multi-stakeholder network to express that social agents of diverse nature (firms, authorities, local communities, expert associations…) may form stable systems to jointly manage complex issues (Gray & Purdy, 2018). It has also been credited that rivalry relations between firms can stimulate the emergence of a new social system with its own contingency structure (Cool & Dierickx, 1993; Kilduff, 2019). This type of agency has also been explored within firms, highlighting the ability of informal groups to compete or collaborate (Allen et al., 2007; An, 2021). All these lines of research have one point in common: they assume and confirm the ability of social systems in business to engage in double-contingency relationships, contributing to the emergence of higher-level systems. In the next section, I will explain how we can identify these systems.
Identifying Social Systems in Business
In his early work, Luhmann (1964, p. 20) referred to organizations as ‘entangled structures’ that combined formal and informal structures. In his view, the informal structure complements the formal structure to enable an organization to adapt to changes in its environment or to overcome internal problems arising from changing and conflicting roles. In the early 1980s, Luhmann changed his approach to organizations, as a result of his new theoretical commitment to social systems as the key notion of his thinking. From then on, Luhmann (2000, 2003) argued that organizations are made up of decisions, which are a special kind of communications. This turn led Luhmann to interpret formal structures as the primary organizational body, neglecting social dynamics taking place within organizations, on their limits or at a supra-organizational level.
In my framework, the relevant point is not the decisions made through the formal structure of an organization, but the interactions that, happening anywhere in an organizational field, generate social systems. From this perspective, the formal structures of any organization play a relevant role, since they provide a prior definition of the expectations to which social agents can adhere when they are about to interact. The top management team is expected to define the firm’s main strategy and communicate it to the department managers. These managers are also expected to collect and evaluate performance data from the individuals they manage. Moving away from Luhmann’s ideas, what is essential in my approach is how and where interactions happen or, to put it more bluntly, how and where patterns of expectations develop through communications. The notion of informal group seems very close to my view of social system, as both describe stable interactions between agents. However, the notion of informal group starts from a limiting assumption: communicative interactions always happen within the formal limits of the firm. In contrast, my notion of social system has a broader scope. To identify social systems, what really count are the communicative flows and the patterns of reciprocal expectations that stabilize those flows, whatever the spatial environment in which they occur. My notion of social system may be applied to those interactions that occur within a firm, between members of that firm and other outside agents, or at a supra-organizational level.
Let us illustrate these possibilities with an example. In a firm dedicated to the sale of machinery for industry, the employees on a sales team interact with each other. This team maintains a competitive relationship with other commercial teams in the firm to achieve the best results. At the same time, the sales team we are focusing on has a very fluid relationship with the design department due to the good informal links between their members. This relationship allows the members of the sales team to quickly solve many of the technical issues posed by customers. In addition, the sales team’s objectives and practices are strongly conditioned by the policies of the sales department and the corporate policies defined by the firm’s top management. Let us imagine that a member of that team arrives at a customer’s facility. At that moment, we could see how the engineer in charge of the factory welcomes the salesman with cordiality. We could recognize two individual actors taking up a previously existing system. Figure 3 illustrates all the social systems that we could identify in this case. In such a situation, it will depend on the interest and perspective of the observer which of these systems deserves priority attention.
Fig. 1
Emergency of a social system. E1: A’s expectations about B. E2: B’s expectations about A. E3: A’s expectations about the expectations that B has formed on A’s behavior. E4: B’s expectations about the expectations that A has formed on B’s behavior
Fig. 2
A social system between physically present and non-present actors
Fig. 3
Identification of the systems involved in a communicative interaction. A: Individual seller. B: Client company engineer. C: Sales team.
D: Rival sales team. E: Technical department. F: Sales department. G: Selling company
×
×
×
So far, I have argued that individuals and social systems can engage as social agents in stable communicative interactions. The result of those interactions will be the emergence of new social systems. I have also claimed that such new systems may emerge in different parts within an organization, in its contact with other outside agents or at a supra-organizational level. From here, I will explore whether such systems may behave as moral agents. To answer this question, I will first reflect on the phenomenon of moral agency. In particular, I will explore the types of moral agency and the conditions needed for each type. I will then assess whether social systems in business meet these conditions.
Conditions for Moral Agency
The Posthumanist Turn
We tend to think of humans as moral beings. Codes may vary between groups, but something in us drives us to judge the behaviors of others and to seek the approval of our reference groups. For centuries, this tendency has justified the theoretical reflection on morality, particularly on who (or what) can be the subject of expectations and responsibilities. Such reflection has traditionally focused on the human being, considered the only entity with the capacity for moral agency. Since the Enlightenment, a particular image of the human being (as a rational, autonomous and intellectually developed being) has been generally accepted as a premise for asserting the moral agency of humans (or, at least, a certain group of humans). During the last decades of the twentieth century, however, a posthumanist turn questioned the premise of the modern human as the only valid prototype for moral agency (Braidotti, 2016). This turn was not intended to deny the morality of humans, but to open the notion of morality to a wide variety of entities that we find beyond that exclusionary image of the modern human.
This posthumanist turn has taken place in many fields of knowledge. In biology, for example, a growing group of authors attributes certain moral aptitudes to some species of animals (Monsó, 2017; Singer, 2009). Among the most prominent authors, primatologist Frans de Waal (1996, 2009, 2013) has convincingly demonstrated that several animal species (such as some types of primates or elephants) have an innate inclination toward social values such as empathy, justice, or solidarity. Based on this evidence, de Waal argues that humans share with other species a strong inclination toward certain behaviors, which have become more complex in the case of humans. Morality assumes more complex modes among humans, as they can use symbolic language to perform tasks that other species cannot—such as codifying norms or subjecting them to discussion and agreement—but morality has a root that humans share with other animal species (de Waal, 2006).
A similar trend has developed in the study of objects and technology. Following Latour’s (1992, 2002, 2009) ideas on morality and objects, several authors argue that objects are not a neutral medium for human action, but often contribute to co-create morally relevant situations, as they expand or limit decision alternatives for humans (Fraser, 2013; Martin, 2019). Without denying the responsibility of humans, these authors claim that moral agency is not located exclusively in the person using an object, but in the combination of both. As an example, Verbeek (2008) analyzes how moral agency is configured when obstetric ultrasound technology is applied. In this essay, Verbeek emphasizes that this technology contributes to creating a specific situation, which links the future parents to their unborn child, and this technology is the basis for the parents to make a moral decision about abortion. In Verbeek’s approach, moral decision-making—and, consequently, moral agency—becomes a joint issue involving human beings and technological artifacts. In Verbeek’s words, “technologies play a fundamentally mediating role in human practices and experiences, and for this reason it can be argued that moral agency is distributed over both humans and technological artifacts” (p. 24). Verbeek deduces from this idea that moral agency expands beyond the specific situation in which actors decide, to affect even the agents who have participated in the design of the technologies.
A Systems Approach to Moral Agency
Unlike other organizational authors, Luhmann explicitly addressed the issue of morality. For Luhmann, morality is a communicative artifact that serves human beings to carry out a complex task: to evaluate to what extent their mutual expectations are compatible. This task can only be performed in a simplified way: using a binary code (good/bad), a structure of expectations and a symbolic index, which is esteem (Luhmann, 1992a, 1996). It is through esteem, its opposite value—disdain—or its zero term—indifference—that an agent can express that, from its point of view, the integration of its perspective with that of other agents is adequate, irrelevant, or incompatible (Luhmann, 1995a). Through the index of esteem (ranging from the positive values of esteem to the negative ones of disdain), a social agent can express to other agents the degree to which it considers the system’s continuity viable and valuable. However, moral communication is not reduced to expressions of esteem. According to Luhmann (1991), moral communication also serves agents to express and coordinate (1) the conditions required to merit esteem and (2) the judgments that, based on those conditions, any agent deserves or would deserve.
From these ideas, I offer my definition of moral agency. In particular, we can attribute moral agency to any human or non-human entity that expresses esteem to other entities of the same or different nature, appears to be sensitive to the esteem of others, and is also capable of engaging in communicative interactions about the norms deemed necessary for esteem and the judgments that, on the basis of those norms, any agent deserves or would deserve. This image includes two groups—or levels—of conditions for an observer to appreciate: (1) the ability to express and react to the esteem of others (I refer to this level as the expressive level of morality) and (2) the ability to communicate with others about the rules and judgments related to esteem (I refer to this as the reflective level of morality). At the expressive level, agents use esteem to express their judgments. At the reflective level, they problematize the rules that are taken for granted at the expressive level. When agents communicate at this level, the rules that coordinate interactions are the object of specific communication. The system’s norms are expressed, questioned, negotiated, and modified by the agents involved in the system. Thus, we can say that the reflective capacity of the agents allows the social system to self-reflect on its normative structure. This view of moral agency assumes the modalities of moral communication identified by Luhmann (1991, 1992a). But unlike him, I dissociate moral agency from human beings and generalize its scope to any human or non-human entity. In my image, moral agency is open to any kind of entity that performs these tasks, including social systems.
This image of moral agency provides an answer to my first research question: what conditions should an entity meet to be recognized as a moral agent? In the next subsection, I draw on my answer to this question to identify three types of moral agency. Having identified these types, I will address my second research question, assessing whether social systems in business meet the conditions I have described.
Types of Moral Agency
My definition of moral agency implies two levels of conditions to be appreciated in an entity: (1) the ability to express and react to the esteem of others and (2) the ability to communicate about the normative structure operating in a social system. The first level helps us to identify the entities that use esteem to coordinate their behaviors. The second level is more restrictive and limits moral agency to those entities that can communicate about the normative structure on which esteem depends. In this image, morality is interpreted as a complex phenomenon in which many entities participate in different ways. In this essay, I focus on the social systems that emerge in business. However, I explicitly acknowledge that other types of entities (human or non-human, known or unknown, existing or yet to be invented) may behave as moral agents.
At this point, I identify three types of moral agency, depending on how an entity participates in the situation. First, an entity can make a difference in a situation without fulfilling any condition for moral agency. In this case, the entity has a relevant influence on a moral situation, but it does not meet the conditions for moral agency. Therefore, we could say that this is an improper case of moral agency. Several philosophers of technology and postphenomenological authors have already suggested this case when they claim that objects and technological devices often play an active role in shaping a situation in which humans must make some moral decisions (see Section “The Posthumanist Turn”). Following Brey (2014), I refer to such entities as moral factors.1 In my framework, a moral factor is a human or non-human entity that makes a difference in a social interaction the observer considers normatively relevant, without that entity expressing or reacting to the esteem of others, or communicating about morality. This would be the case of a traffic light, which normatively influences a situation by declaring when drivers should brake. Many immaterial objects—such as cultures, fictional characters or ideologies—can also participate in a situation as moral factors. This would be the case of pro-violent messages uploaded to the Internet by a fundamentalist religious group, influencing individuals thousands of miles away. In any of these cases, the moral factor makes a difference in a situation that appears to be morally relevant. We would criticize the driver for running the traffic light or those who posted the messages inciting terrorism.
Second, we can identify those entities that only meet the first set of conditions for moral agency. In particular, I define an expressive moral agent as an entity that exercises the appropriate ability to express esteem and react to the esteem of others. As mentioned above, some biologists and philosophers have already attributed this type of agency to certain species of animals, such as killer whales, dogs, rats, elephants, and primates (de Waal, 2009, 2013; Monsó, 2017; Singer, 2009). I referred to such recognition in Section “The Posthumanist Turn” of this essay.
Third, we can identify those entities that fulfill both levels of conditions. In particular, I define a reflective moral agent as an entity that exercises the appropriate abilities to express esteem and react to the esteem expressions of others, and is also capable of engaging in communicative interactions about the rules that are deemed necessary for esteem in a social system and the judgments that, on the basis of those rules, any agent deserves or would deserve. In my framework, moral agency is an open and heterogeneous category, which can be attributed in different modalities to any entity that fulfills certain functional conditions from an observer’s perspective.
At this point, it only remains to assess whether the social systems that emerge in business (see Section “Identifying Social Systems in Business”) might be recognized as moral agents of those types. I will address this issue in the next section. Before that, I summarize the conditions for moral agency and add some examples in Table 1.
Table 1
Types of moral agency
Conditions
Examples
Moral factor
An entity makes a difference in interactions connected to an issue that the observer considers normatively relevant
That entity does not express or react to esteem, nor does it communicate about moral rules and judgments
Ultrasound test allows parents to make decisions about the life of their unborn child
A traffic light prompts drivers to stop to yield to pedestrians
A parent’s advice guides a young man to make a morally relevant decision
Expressive moral agency
An entity expresses esteem and reacts to the esteem of others
That entity does not communicate about moral rules and judgments
A jailed monkey protests receiving a lower reward than a peer during an experiment
A rat prefers to release another rat from its confinement rather than access food
A pod of killer whales leaves a piece of the hunt for an elderly member
Reflective moral agency
An entity expresses esteem and reacts to the esteem of others
In addition, that entity communicates effectively with others about moral rules and judgments
A group of activists for the planet demands immediate effective measures in the city, where a G20 meeting is taking place
A team’s coach asks supporters to cheer for their team with an attitude of respect for the opponent
The CEO of a large company speaks out publicly in favor of migrants’ social rights
Conditions and examples
Social Systems in Business as Moral Agents
In the previous section, I identified three types of moral agency depending on the functional conditions met by the entity. In this section, I explore whether social systems in business can assume these types of agency.
Social Systems in Business as Moral Factors
As mentioned above, an entity can be recognized as a moral factor when it makes a difference in interactions that the observer considers morally relevant without that entity expressing or reacting to the esteem of others or communicating about morality. We can recognize as moral factors certain material objects, such as a subway turnstile or a defibrillator machine hanging on a wall, but also immaterial objects, such as the ideology of a political party or the instructions for interrogating prisoners used by an army. The mere existence of these objects makes a difference in a situation, inhibiting certain behaviors and promoting others that are evaluable from a moral perspective.
However, can we recognize social systems—particularly those emerging in business—as moral factors? From a theoretical point of view, a social system may operate as a moral factor in two alternative ways: (1) by influencing the actions of the agents involved in the system and (2) by influencing the behaviors of agents not involved in the system. The first case would occur, for example, when a company’s culture induces an employee to treat a customer with attention and kindness, or when a factory worker adjusts his effort to the norms of his informal group. The second case would occur, for example, when a top management team designs the company’s environmental policy taking into account the possible reaction of the most belligerent NGO. In these and many other similar cases, a social system (formal or informal, internal or external to the firm) influences the behavior of other agents (individuals or social systems) in a morally relevant situation, even if there is no direct interaction between them.
Organizational authors have highlighted the first type of influence—from a social system toward agents involved in the system—in diverse situations. To mention some examples, Paring and Pezé (2022) conducted ethnographic research in a financial corporation to show how a group of individual consultants make normatively relevant decisions that are reactively influenced by the firm’s management policies. In a different vein, Schauster et al. (2021) showed how advertising professionals make morally evaluable decisions by conforming to the norms of their professional reference groups. As another prominent example, Falkenberg and Herremans (1995) found that informal systems exert a relevant influence on individuals when they have to deal with morally complex issues.
The second type of influence—from a social system toward agents not involved in the system—has been widely highlighted from the stakeholder approach. In this line of research, Freeman et al. (2007) assert that managers of any firm should make decisions taking into account the interests of other agents (customers, investors, local communities…), including those agents that receive the effects of the firm’s decisions without being linked to it in a direct interaction. Similarly, Wang et al. (2016) highlight the responsibility of firms toward society, interpreting society as a collective system that receives the effects of firms’ decisions. Starting from the notion of common good, Sison and Fontrodona (2013) emphasize the responsibility of firms to provide goods and services that meet the needs of a broad group of agents that operate around them. Turning their attention to different parts of business, all these authors emphasize the ability of corporations—or their management teams—to consider the potential effects of their decisions on other social systems, being those systems acting as moral factors in relation to corporations or their management teams.
Social Systems in Business as Expressive Moral Agents
An expressive moral agent is an entity capable of using expressions of esteem to coordinate its expectations with other agents. We often attribute this ability to social systems in business. We do so, for example, when we see two sales teams competing to obtain the best results, or when a management team reacts to pressure exerted by a group of activists to change a corporate policy. In situations of this kind, we accept that social systems participate with other agents in double-contingency relationships and are able to use expressions of esteem to coordinate their mutual expectations and behaviors.
From a theoretical point of view, social systems cannot express esteem or react to the esteem of others in a phenomenological way. They cannot smile, hug, or protest as individuals do. Nor do they have consciousness, intentions, or emotions, and any assignment of such qualities is always done in a derivative manner. In my view, however, the condition for expressive moral agency is not the phenomenological exercise of an internal ability that is connected to embodied actions (such as speaking, looking, touching…), but the plausibility of attributing to a system the ability to coordinate expectations through esteem. What really counts is whether a social system seems capable of expressing its positive or negative evaluation of the continuity of its interaction with others. In the case of the social systems in business, this ability is always exercised in a mediated way, that is, through the individuals or texts speaking on their behalf.
The recent LGBTQ + rights controversy involving Disney provides a good example of this type of agency for social systems in business. In March 2022, a broad range of LGBTQ + groups and activists called on Disney to use its influence to prevent the Florida Congress from passing a bill (HB 1557) popularly known as Don’t Say Gay. With a majority in Congress, conservative politicians wanted to restrict the teaching in schools of issues related to sexual orientation and gender identity. In reaction to these pressures, Walt Disney CEO Bob Chapek sent an email to all company employees. In that email, Chapek defended Disney’s neutrality. In Chapek’s words, “corporate statements do very little to change outcomes or minds. Instead, they are often weaponized by one side or the other to further divide and inflame” (Mike, 2022). By arguing this way, Chapek defended himself and Disney from previous accusations that were interpretable in terms of esteem, and avoided a conflict that he considered counterproductive.
In the days following that email, Chapek and Disney were again heavily criticized by LGBTQ + employee groups and outside activists. In reaction to that criticism, Chapek sent a new message on behalf of Disney to publicly apologize for his silence and express that Disney would actively oppose the controversial bill (Pallota, 2022). In his email to Disney employees, Chapek said, “To my fellow colleagues, but especially our LGBTQ + community. Thank you to all who have reached out to me sharing your pain, frustration and sadness over the company’s response to the Florida ‘Don’t Say Gay’ bill […]. Our employees see the power of this great company as an opportunity to do good. I agree […], we need to use our influence to promote that good by telling inclusive stories, but also by standing up for the rights of all” (Disney, 2022). He also announced that Disney would be “pausing all political donations in the state of Florida” (Disney, 2022).
Once the commitment was announced, Chapek and Disney were again criticized, this time by other employee groups, outside conservative groups, and Florida politicians (Durkee, 2022). The Disney case shows that moral dynamics in business are often driven by a wide range of social agents (individuals, management teams, employee groups, politicians, external collectives…) that engage in relationships and coordinate their expectations through esteem. This coordination is only possible to the extent that these agents are functionally able to express and react to the esteem of others. An important point is that these agents, when communicating, take for granted the normative structure that justifies esteem. Speaking for Disney, Chapek expressed esteem to some agents and reacted to others’ expressions, without discussing the norms or expectations the company should meet to merit esteem.
In the organizational literature, the ability of social systems to express and react to esteem has been supported by several lines of research. In particular, the notion of moral legitimacy (Suchman, 1995) has served many authors to attribute this ability to corporations. Although this notion initially emphasized a firm’s efforts to conform to society at large (Dowling & Pfeffer, 1975), it is now usually interpreted as a resource or strategy that allows the firm to gain the support of specific stakeholder groups. In this approach, Bosse et al. (2009) and Fassin (2012) emphasize that firms engage with other agents in relationships that involve the practice of moral values such as commitment, mutual support, and loyalty. Other lines of research have also supported the expressive moral agency of social systems by examining intra-firm dynamics. Tchokogué et al. (2017) explain that managers in the purchasing department of a firm can reinforce the legitimacy of their department, thus influencing the interactions with other departments and the firm’s top management. LaVan and Martin (2008) explain that informal groups play a relevant role in the dynamics of employee bullying occurring within firms. Assuming different perspectives and focal points, all these lines highlight the ability of social systems in business to coordinate their interactions through esteem.
Social Systems in Business as Reflective Moral Agents
So far, I have explained that social systems that emerge in business can behave as moral factors or expressive moral agents. I must now assess whether these systems can also be considered reflective moral agents. To do so, I must take into account the definition of this type of agency I offered above. According to that definition, an entity behaves as a reflective moral agent when it is able to manage esteem in its interactions with others (first level of conditions) and is also able to communicate about the esteem-related rules and judgments that are applicable in a system (second level of conditions). The second level of conditions sets a more restrictive perimeter for identifying reflective moral agents.
When agents operate at this level, the structure of norms and expectations that were taken for granted at the expressive level becomes an object for communication. Norms and expectations are now thematized. By operating in this way, communication between agents allows the social system to self-reflect on its own normative structure. When one young man learns that his friend has revealed something that he had previously told him as a secret, he decides not to express his anger. Putting that reactive attitude aside, he explains to his friend how important it is to keep secrets private when a relationship is based on trust and respect. The friend responds by arguing that such a rule may be valid when a secret is told to people who are completely unrelated to the couple, but not when it is told to people who have a sincere interest in helping and protecting. Beyond any direct expression of esteem, friends are communicating about certain norms that are taken for granted at the expressive level of morality. This reflection opens up the possibility for agents to negotiate the validity, and the normative force, of the rules and expectations operating in the system.2
When we look at business, we usually attribute to social systems the ability to communicate about the structure of norms and expectations. We do so, for example, when we read about a management team negotiating with an activist group to modify the corporation’s environmental policies, or when we hear about a worker speaking on behalf of his team to explain to the newly hired employee how they coordinate their tasks in the factory. In these and many other similar cases, we readily accept that social systems (a management team, an external activist group, a team of employees…) have the appropriate aptitude to communicate with other agents about the structure of norms and expectations operating in the relationship. Although this type of agency would be denied by individualist authors (see Section “Communication as a Peripheral Phenomenon to Moral Agency”), it is easily acceptable if we assume a functionalist approach and focus on the systems that emerge through communication.
Throughout 2021 and 2022, Amazon was involved in a similar controversy to Disney, but the communication dynamics happened differently. In February 2021, Amazon stopped selling the book ‘When Harry Became Sally,’ considered transphobic by the LGBTQ + community. As a result of this decision, Amazon’s CEO received a letter from four Republican senators accusing the company of censorship (Rubio et al., 2021). Amazon’s CEO responded to this letter by sending another to several Republican senators stating, “we have chosen not to sell books that frame LGBTQ + identity as a mental illness” (WSJ Staff, 2021). Just a month later, several hundred employees signed a petition asking the firm to withdraw the book ‘Irreversible harm’ because, in their opinion, it violated the firm’s policy on selling books that frame LGBTQ + identity as a mental illness. On this occasion, the top management refused, justifying the decision by Amazon’s policy of providing access to diverse viewpoints, “even when those viewpoints differ from […] Amazon’s stated positions” (Kim, 2022). As a result of that conflict, representatives of Glamazon—an affinity group created by Amazon to give voice to employees of the LGBTQ + community—held several meetings with Amazon executives. As part of the discussion, Glamazon representatives demanded “to change the book’s category to ‘conservatism’ instead of ‘transgender rights & expression’, label the book as misinformation, donate the book’s proceeds to a trans-focused charity, and provide more transparency into the decision process” (Kim, 2022). The management team expressed sympathy for the collective, but refused to carry out those actions.
In June 2022, a group of about thirty Amazon employees stormed an event organized by the corporation next to its Seattle headquarters. At the cue of a siren, the activists lay on the ground, interrupting the speech of Glamazon leaders. As a result of that protest, the event was canceled (Rosenblatt, 2022). The activists were members of a new group, No hate at Amazon, which has recently emerged among Amazon employees, without the initiative and supervision of the firm’s senior management. During the protest, the group demanded that Amazon stop selling books they considered transphobic. In the words of one of the protest organizers, “We believe in free speech. We believe in a free marketplace of ideas […], but we draw a line against hate speech” (Rosenblatt, 2022).
The Disney and Amazon cases are related to a similar issue, and even involve the same types of agents (management teams, politicians, employee groups, external collectives…). However, the communicative dynamics happened very differently. In the Disney case, the dynamics operated mainly at the level of expressive moral agency. Agents expressed their valuations of Disney’s actions and omissions, taking moral rules and expectations for granted. Agents expressed and reacted to esteem. In the case of Amazon, the dynamics also occurred at the reflective level. Social agents were striving to communicate about the rules that should guide Amazon’s actions and, with a broader scope, were expressing their normative views about this kind of behaviors. In both cases, the communicative dynamics were actively developed by certain social systems (management teams, formal and informal internal groups, external collectives…) that have not traditionally been recognized as moral agents in business (see Section “Communication as a Peripheral Phenomenon to Moral Agency”).
In the organizational literature, several lines of research have indirectly supported the reflective moral agency of social systems. As an example, Scherer, Palazzo, and colleagues have extensively explained that firms and other social agents engage in large communicative systems to jointly negotiate the norms and goals that will guide their behaviors concerning an issue (Scherer & Palazzo, 2007; Scherer & Voegtlin, 2020; Voegtlin et al., 2012). Other authors have explored the dynamics of dialog and coordination between a firm and one of its stakeholders on issues that both consider normatively relevant. Burchell and Cook (2013) show how reflective communication between firms and NGOs can lead to transformations that improve mutual engagement. Olabisi et al. (2019) examine the transition of an indigenous community from its position as a non-stakeholder to the status of a primary stakeholder for a firm, acquiring legitimacy that allows the community to voice its opinion on certain structural elements of the interaction. In another vein, Hill and Rapp (2014) look inside the corporation to offer a bottom-up image of the dialogic process that should result in a firm’s code of ethics. With a theoretical approach, these authors assert that the development and implementation of a meaningful code of conduct should emerge as the result of a collective, participatory, bottom-up reflection that involves several discussion teams within the firm. In all these works, the authors support the idea that social systems participate in stable relationships in which they discuss and negotiate the norms and judgments that affect their behaviors, in relation to issues that the agents consider normatively relevant.
Final Discussion
In this essay, I have presented a systems approach to moral agency in business. In my view, moral agency is not a homogeneous phenomenon and includes three modalities according to the functions played by the entity: moral factor (the entity makes a difference in a situation without actively participating in the communicative interactions), expressive moral agency (the entity expresses and reacts to the esteem of others), and reflective moral agency (the entity communicates about esteem-related norms or judgments). The essential criterion for these types is the role played by an entity in a communicative situation, that role being appreciated by an observer.
This approach does not presuppose that moral agents are internally inclined toward goodness, dialog, or agreement, nor that they are rational or sentient beings. I interpret morality with a functional lens that attends to the roles played by agents in communicative interactions. Such interactions can be oriented to collaboration or agreement, but also to conflict. Moreover, this approach goes far beyond the restrictive boundaries of anthropocentrism by opening morality to a wide repertoire of human and non-human entities. In this essay, I have articulated this general framework and applied it to the social systems that emerge in business.
My approach is far from other theoretical proposals on moral agency in business. In my view, moral agency depends neither on the intrinsic qualities of the actor nor on the utility to society (see Section “Communication as a Peripheral Phenomenon to Moral Agency”). In particular, I regard moral agency as a complex phenomenon, which can only be explained by taking into account the specific functions that an entity performs in a particular interaction, these functions being appreciated by an observer. At the same time, I consider that moral agency does not pre-exist communication, but emerges in it (see Section “Communication as a Central Phenomenon to Moral Agency”). Social and moral agencies are not attributes that we assign to entities that fully exist before they concur in interactions. On the contrary, both types of agency describe how an entity emerges into existence when it participates in interactions. An entity becomes a social agent when it contributes to the emergence of sociality, and it becomes a moral agent (in any of its modalities) when it contributes to moral dynamics.
My approach is based on Luhmann’s ideas on social systems and morality. However, I have introduced several relevant modifications to his ideas. In relation to his theory of social systems, I have made three relevant changes. First, my image of those systems attends to all flows of communication, even those occurring in a diachronic or mediated way (Berglez & Hedenmo, 2023). Second, instead of assuming that only individuals perform communication, I assume a functional approach that allows me to offer a more open and heterogeneous image of agency (de Waal, 2006; Jackson & Williams, 2021). Third, I take the notion of social system as the basic unit to give an account of the communicative dynamics in business, leaving aside Luhmann’s preference for formal organizations (LaVan & Martin, 2008; Scherer & Voegtlin, 2020). In relation to morality, I have departed from Luhmann’s logic on one relevant point. In Luhmann’s theory, morality is a phenomenon restricted to humans. Although Luhmann displaces humans from their privileged place in social theory, he still believes that humans are the only plausible entities for sociality and morality. I do not start from this premise. In my view, any entity that performs the appropriate functions should be considered a social or moral agent (Monsó, 2017; Verbeek, 2008).
At this point, it may be helpful to recall my epistemological approach. Specifically, I construct an image of social systems and morality and then identify what conditions an observer should appreciate in order to rigorously attribute moral agency to a system. As discussed in Section “Emergence of Social Systems,” social reality is incomprehensibly complex, and can only be accessed through observation. For this reason, various observers may interpret differently the communicative dynamics occurring in a social context. Several observers may identify and focus their attention on different social systems (see Section “Identifying Social Systems in Business”). This does not mean lack of rigor or arbitrariness. If an observer’s interpretation is justified by a functional analysis such as the one I propose, those interpretations should be acceptable, at least to that observer.
Finally, I would like to point out some lines for future research. This essay assumes a theoretical approach to explore the attribution of moral agency to social systems. Therefore, this proposal might be complemented with a field methodology that allows us to identify the social systems involved in a situation. This methodology should be similar to the one that Cooren and other Montreal School authors often apply to explain agency in communicative situations (e.g., Brummans, 2018; Cooren, 2007; Taylor et al., 2021). Such an approach would help us address many interesting questions, such as: how actors negotiate the attribution of moral agency, how agency allocation influences interactions, or what factors facilitate or inhibit a system’s willingness to assume agency-related responsibilities. The notion of moral agency refers to the expectations and responsibilities addressed to an entity. In fact, the issue of moral responsibility becomes relevant when some kind of controversy or scandal occurs. For this reason, it would be convenient to explore the issue of responsibility in business when considering social systems. Thus, it would be useful to investigate what conditions should be fulfilled to consider a social system as morally responsible for an event, or how we should distribute moral responsibility among the agents involved, whether they are individuals, firms, or infra- or supra-organizational social systems. This essay may also stimulate relevant research beyond organizational studies. In particular, its functional approach to agency could be projected to other non-human entities, such as animals, robots, AI systems, or hybrid networks of human and non-human elements. In addition, I have made some simplifications that could inspire other theoretical developments. The logic of social systems could stimulate authors to explore questions related to the distribution of agency in complex situations. This would be the case of tensions between actors in a system, observers with different symbolic power or the temporal evolution of perceptions of agency.
Conclusion
Moral agency is a relevant topic in business ethics. In an increasingly complex world, we need to reflect on who or what we consider to be the subject of expectations and responsibilities in business. Much has been written on this question without reaching a basic consensus. In general, the dispute remains between those who point to individuals or firms as moral agents. My essay approaches this question from an alternative perspective. Specifically, I interpret moral agency with a functional approach based on communication and identify three modalities (moral factor, expressive moral agency, and reflective moral agency). This essay extends a communicative turn that has been initiated in theoretical research on moral agency in business. Hopefully, it will serve as a basis for fruitful developments in business studies and other fields of knowledge.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Although Brey (2014) admits the moral agency of non-human actors, he draws an ontological barrier between objects and humans. According to Brey, only human beings can be moral agents—for only they can act intentionally—and objects only participate in morality in a derivative way. I accept his proposal to designate objects as moral factors, but I disagree with his essentialist and anthropocentric image, far from my functional and ontologically open approach.
Luhmann repeatedly recognized this capacity of social systems to reflect on their own structures through the agents’ communications (1995a, p. 45, 132). More specifically, Luhmann (2008, p. 112) recognized that moral communication could become self-reflexive, serving the agents involved in a system to control the risks derived from their expressions of esteem.